Factor Investing | Institutional Masterclass

  • |
  • 46 mins 23 secs

There's a growing interest in factor investing but is it a passive or active investment strategy? Topics also discussed include how investors can blend factors in a portfolio and crowding out returns. On the panel are:

  • David Schofield, President, Intech International Division, Intech Janus Henderson
  • Ashley Lester, Global Head of Multi Asset & Portfolio Solutions Research, Schroders
  • Michael Kinney, Principal, Mercer



- Automatically Generated -

There's a growing interest in factor investing, but is it a passive or active investment strategy ? I'm marco gate. This is institutional masterclass. To discuss the topic. I'm joined in the studio by david scofield. He is president of the in tech international division, part of in tech. Janice henderson. Actually lester, head of multi asset research, and michael kinney, principal at marcia. David scofield how long has intact being involved in factor or factor type investors ? Well, it must have been controversial actually were not involved, in fact, investing a tool, but we i care deeply about it, we think about it. We're interested in what drives the behavior of portfolios. We've been involved in quantitative portfolio management ourselves since nineteen eighty seven, so just over thirty years now we don't describe what we do is factor investing. However, the approach involves alternative waiting schemes to cap weighted indexation, which obviously has a lot in common with other alternative approaches to building equity portfolios such a smart beetle factory and what's the view that schroeder's what what do you classify as a factor ? How long have you been doing it ? So schroeder's, we take a holistic approach to factor investing in the sense that it expresses a systematic way of undertaking investment across potentially any asset class, or in a multi asset class context and that's actually, what cut short has started. In fact, you're investing in the wake of the financial crisis we thought instead of the heat breaking down our investments across asset classes per se, we should look at the underlying wrist drivers, which equivalently could be thought ofthis factors so schroeder's multi asset has been undertaking factor investing across the range of asset classes since shortly after the financial crisis. So could you give us example of what some of those factors might be ? Our they classic ones like low volatility momentum value are really actually, this is a really fascinating topic when we think about factory investing as applied teo asset classes beyond equities often factor investing his thought off purely with regard to equities, and i suspect that's where well, mainly talk about today. But there's, a rich sort of set of strategies which one can think of a cross asset classes and within across asset class, is kind of the classic example is trend, which is like a time siri's momentum way of undertaking allocation across asset classes within asset classes. There's quite a quite a rich sit off possible strategies, which looked quite different from equity strategies, and the classic is to play the slope of the yield curve, or the slope of the curve of the implied ball curve those of varieties of carrie strategies, which have no very ready equivalent within equity. Investment brought a broad box of tools that michael, how do you look at it at mercer ? Do you incorporate it in how you judge fund managers, or do you judge for managers who happened to run five models ? Okay, so our role at mercer assed to help people build robust global equity portfolios ? If we stick with inequity on thinking about the factor exposures, that one has a very important part of that little very pie client, depending on their goals and constraints, what it is they're trying to achieve, but the consideration warp factors aaron, a portfolio portfolio structure very important. We've been doing that for explicitly for for some time off. Since twenty ten david, you are making a clear distinction between what counts is quant and factor. Could you just develop that a bit more ? For sure, our starting point is the acknowledgement that cap weighted portfolios are not necessarily the most efficient way to allocate cattle in equity investments. So trying to come up with some kind of systematic, not necessarily rules based, but often some of these bridges are rules based approach to re waiting the cap plated portfolio away from cat waiting on. Dh then maintaining that more efficient structure through time, which depends, of course, on trading and re balancing. And so this is something which we spend a lot of time researching the implications of rebalancing portfolios alternatively, waited portfolios. So i think the sort of active quant world has been in many ways, reduced to the sort of factor approaches that we've been hearing about in recent in recent years, however, there are more than one way to skin the cat, so it does make it so, i think it's not important not to get to focus exclusively on factors which seems to be happening in the industry at the moment, it seems to be dominating, dominating the discussion, michael picking up that wise factors such a buzzword of the moment while we seeing so much interesting, we always have had a lot of interest, but we haven't always used the buzzword factors halfway, so so the label is new, and perhaps the underlying issue has been around with us for much, much longer. So but i think the availability of the tools and techniques that we have to perhaps understand those factors in greater, greater granularity has brought it. More to the foreign in recent years. But but it is something that it's not it's, not a new topic. It's a new word and would you say factor invested marchal is something that's, active or passive ? How do you work out ? Try to avoid the trite afford to putting those labels are onto factor investing, but no, no factor investing requires its complex. It requires tough decisions being made it's not a guarantee of success, and it requires active decisions. Eso factor investing, which i think of a systematic rules based approach to target certain risk premium in a transparent and low cost way, is a form of active management. I should beat you. I would absolutely agree with michael's view on factor investing as active management. I think there's been a favorite of confusion around this and passive has some sort of natural attractions, perhaps in the sense of providing a safety blanket in a sense of a lack of decisions that require to be made bye and investors, but ultimately quite fine grained decisions within factor investing can have an enormous effect on your portfolio. A very simple example of this is if we look at some of the best known index providers and just looked within a given factor implemented on a given index. So, for instance, three different nsc i value siri's the accumulative returns about the index over the last twenty years have varied between plus ten percent. On the one hand, on more than plus one hundred twenty percent on the ella, you know, that's a massive, massive, cumulative difference. Those decisions are all within what the industry conveniently chooses to call value factor. So decisions are at the heart of it. And that means david yeah, i agree entirely, actually. With, well, actually just said on one of the difficulties, i think facing investors has been this proliferation of factor based products that is out there way see hundreds of factors that have claimed to have been identified by academia to a greater or lesser extent on dh not to mention index is based on those factors there. Arm or index is now in the world that there are stocks which makes sort of choosing the correct in next if there is such one to meet your needs as an investor becomes a very all active decision, you see the same sort of dispersion of returns, for example, in the low volatility space, you have indexes from major index providers purporting to do the same thing. And yet with annual differences in return of ten, fifteen, twenty percent it's very definitely an active choice that investors need to make and not an easy one, but that's, the choice they infest makes on which product they picked. But when you say that these are actively managed, is that because you've actively set up a system for running money that's different from a market cap weighted index, or is it because even within having set that systematic system up there is still points where human beings have to get involved to make individual decision it's very much the former ? So i think all three of us here talking about strictly systematic approach is to actually running a portfolio on today. So the decision making in factor investing from the asset managers perspective comes into play in how we set up the rules or the algorithms which determine what investments will be made. Overwhelmingly the algorithm or the rules determine which stocks actually end up being held. I'd like to pick up, though, on a point that david was making about this proliferation of factor ideas, i think, is a convenient consensus in the industry that everyone knows that there's say five non market factors size, momentum, value, quality level. Come on, then, there's on the other end of the sort of spectrum, the three hundred or four hundred and ever evolving number off factor ideas or signals which academics have proven in some sense to work. I think what this really illustrates is again another dimension of the choices they're involved in factor investing. Not all of those three hundred different signals are actually different themes. Makes sense to sort of group quite a lot of them within value, for example, but which choices you make within value and how many off them happen ? The group in value again, that's an active decision informing the algorithm by the asset manager. And it has a huge impact potentially on the success of your strategy. Well, michael, how do you distinguish between what you think is a fact with legs rather than something fashionable ? With a record, that's worked up to the point. You put client money ? Sure. Ok. Bye. Asking three questions when looking any hypothesis around a factor on de so merciful every time we've looked at gamut of factors out there in salt toe identify which ones were well counters a unique factor. Those three questions are firstly. Is there rigorous, peer reviewed academic work ? That's been published that is supportive of that fact. But that's just the first one. The second is then to do our own quantitative work by by looking at how is performed in the past and doing a number of statistical tests. So that's our own work on the factors well and then the third thing is there a narrative, a causal reason for the future ? For wide that particular risk, premium or factor should work. And in the future you could be a behavioral or it could be economic rationale. So we look at those three three questions than the holistic question that then one adds to all of those. Those three early questions is are they orthogonal factors different on trying the energy they complement each other all day ? Just a different way of saying the same thing. And can i ask you a question about that's ? What sort of commonality of underlying exposes do you ? Find between the big factors sort of fact, of all the things i think would begin that depends on which particular metrics one uses to define a factor, but i think you'll find a fast majority commercially available products, not perfectly a thorn on there is an overlap, yes, but the reason i ask the question because i think we think back to the last sort of financial crisis, the so called quantum meltdown that occurred second half of two thousand seven and continued through two thousand eight what i think was potentially exposed their wass sort of commonality between products that were otherwise thought to be different, because it turned out that when when the crisis actually did hit, that there was quite a lot of commonality, there's quite a high correlation. In fact, underlying these different quantitative products, which were all essentially doing variation on a similar theme of sort of multi factor type tilts we're picking out actually part of that, that is, when you build your own product, if you build it in isolation, you don't know whether what you're producing is a work of genius or something everybody next you know, all the shops up. The road is done, i mean that's the value of having things like trade wires, that question arrive, and so i agree that a sensible starting point one reason why factor investing is so successful is that it provides grand rules for scientific discussion, about different approaches to investment. And so that's how we've grown from one factor in nineteen sixty for the market for hundred factors today, so it provides the ground rules for progressive discussion, and in conducting your own research, you'd better look around at what else is out there, both in academia and in industry and to michael's list kind of criteria, i would add that from our perspective, a very important criterion is can refine factors in which we believe as much, for instance, as a very widely used factor, but which were reasonably cauldron isn't quite as widely used. It doesn't need to be some magic nobel prize winning secret source, but something which pushes you off a bit kind of that the path off the highest capacity products, which are most likely to be kind of having a way. And so two examples off so very commoditized factors book two prizes value, which is perform hideously over the last ten years when lots of other measures of family you have done fine and just the rule last twelve minus one. Month's returns as a measure of momentum, which has done fine but has some very undesirable characteristics, including a tendency to crash it inopportune times. And there are other measures which are reasonably well known, which can push you away from that obvious and simple example, being at least control for the volatility of the underlying stocks. When you're measuring the momentum. So always been looking that rich universe and that the universe, which is getting richer gives us an opportunity to diversify away. I am a mental defective, for instance, the holdings in armament invicta around me, about fifty percent correlated with the holdings in a raw standard commoditized momentum factor. You can never be completely confident in investment, but that gives you some extra confidence that were part of a ritual ecosystem now than we were during the crash of two thousand. Well, let's, move on now to the second part of our program should investors look to buy a multi factor fund or put together all the different factor strategies themselves ? Michael what's there's quite a few things to consider there, but it's a it's a good question on the very same sure, my colleague panel colleagues would would agree that the very first thing to think about is the clients and what they're trying to achieve on their goals and objectives are on as well. No acid owners around the world are in different circumstances. They'll have different governments, budgets, different fi sensitivities, different time horizons on dh that'll lead him to come to different conclusions about which factors to gain exposure to so it's, not really a simple is multi factor or single factor. And then a separate question is how do you combine in which our suspect, but so it's a little bit more complex, that one or the other on, of course, the other elephant in the room is, should you be using a systematic approach to achieve exposure to factors in first place ? There is traditional conventional active management judgment, elective management on so manufacture investing is not a choice between one factual, the other, if it's a question of how to gain explosion to it, and traditional active management certainly has a role as well. What ? David, when she was saying that little about how rich the data set now is competitive ten years ago. Twenty but isn't this all getting just terribly complicated away from the world where you simply buy low and sell high well, that z that's a very good question on i guess the answer is yes, and i think there's a danger off of of over complicating things it there was a recent paper a few years ago in which this was done by rob arnott and co, in which they looked at very, very wide range of different waiting schemes, all the traditional factors, a variety of different, other systematic waiting schemes. They took some even some random waiting schemes when they took the inverse of those waiting schemes, and and i was the results of all these portfolios, and it turned out they all outperformed. They will not perform the market. But first blush seems extraordinary, because, of course, active managers seem to be, it seemed to seem to struggle to outperform the's cap weighted index is whereas all of these systematic waiting seems a test in this paper and they're in verses all of which have performed some by more than the original wait on dh the analysis that was done was that it all boiled answer that all of these different waiting schemes all boiled after having a size and a value exposure on this essentially therefore explained everything and that was sort of thoughts to be enough. Yeah, i need go no further wants the factor explanation has been distilled out of the of the results way prefer to go a bit further and actually you're you're sort of bile. Oh so high reference there is very opposite because it turns out that you, khun khun, further break down the reason for the performance of these portfolios for those factors down to the re balancing activity that is necessary to maintain the various different factor explosions in the first place. This certainly applies very strong way to size and value and low volatility that actually the the maintenance of the portfolio, the fact that stocks change their characteristics and therefore you need to replace ones. That no longer fit your factor criteria with others that now do. That sort of trading activity actually explains the majority of the returns. So actually does boil down to a simple bilos so high, because that's sort of re bouncing has that sort of characteristics there's bad renouncing, which is she's selo by that happens too. So that's not so good. Yes, it has had to make money if you if you tell on high but at eye run research, actually strong supports what you finding, david. So we were looking recently. And we, for instance, tend to turn over reports on the monthly many of the sort of commission available. Indices turn over a six monthly. And so we thought, well, what's the cost of that in terms of the expected factor exposure in return on my sort of theory going into that was that it would be pretty costly in terms of momentum exposure, but not so much for the other factors. In fact, that was wrong for momento value and quality. All three of them going from monthly to six monthly. Re balancing you lose on the order of a third to forty percent off your effective factor exposure and so this just goes back again to the point that all of these elements off product design or investment design, active choices and all of them could make a big difference if the best case it's hard enough to come up with a good signal, if you then waste about forty percent of the signal because you don't turn over frequently enough to capture it, then really, the upside is pretty limited. Andi, i just wantto on that point kind of pick up on michael's point about multi versus single factor as well. From our perspective, the dirty secret of the investment industry is that when you talk to quantum about this there's, just almost no question, why would you throw away a bunch of information and invest in one factor of time ? It's just not a sensible thing to do, because then let's say you believe in value and momentum and you hooked by a family court for more than half the stocks in your valued portfolio. Gonna be recent loses that's why they're cheap. So you're degrading your own exposure to affect that you've said you believe in, and that empirical effect is really large for four roughly orthogonal vector's you lose about forty percent of your risk adjusted returns by going one factor a time rather than combining the stock will. So when you look at an individual stock, how many factors could simultaneously be happening in it ? So same time ? Because the reason i asked as well as i confused, i think it was rustled its research and settle when we've run our multi manager portfolios. If we've got value and growth style active managers, the stocks that do the best are the ones that somebody who's a value manager on the growth manager at the same time both hold you say something magic about around our own research absolutely supports that conclusion. So we've got research forthcoming in the journal of portfolio management, which shows that the stocks that do best our stocks that will across all of the signals that you're looking for a traditional, of course, on those signals doing well. But so long as the signals are doing well, the stocks that do best over time in the stocks that do well across a lot of signals and you've got to come up with very funny theories of the world and you sometimes hear these sort of speculated on in order to get away from that basic insight that stocks that do well on a number of characterised six do better than stocks that do well on only one characteristic. And by the way, there's no such thing as a single factor portfolio anyway, because when you buy a value for for you, it should be called longer than it should be called long value short quality short momentum. We've done some recent analysis on kind of well known value in two cities over the last ten years which have suffered terribly. Finding is that value could say in those indices has been basically flat over the last ten years. What's hope those indices is that large, uncontrolled exposures, negative to quality and momentum, which is great. Michael, given all this complexity, what how do you analyze the portfolios ? The managers that i put in front of you ? How transparent are they ? Does your analysis ? Transparency is a prerequisite, teo low cost and transparency to the key features that we're looking for, honey, multi factor or single factor a strategy so that the prerequisites teo possibility of forming a positive view on them. Actually, we look at the way we look at all these poor, their form of long only equity management. On we think of that universe onda multi factor strategy, proprietary managed multi factor strategy sits alongside many other forms of global equity management. And so we look at that's the context, the peer group for looking at them. Sure, there are sort of subsets of the pay group, but the broader landscapers of global equity manages on dh. So in that sense, mercer approaches multi factor systematic managers in the same way that we would do for any other actively managed fundamental strategy versus established for factor, framework of idea generation, forfeited construction implementation, business management. But getting in the context is important to remember that active management is an alternate what's going on what's left for active management. If all of this intellectual power this computer firepower, this ability to analyse everything takes what's left for the poor active manager. Sure, there will be a fantastic question. Actually, vash is actively colleagues who ran actively managed to get some within a minute. But that failed in that. I think, you know is you said you know, active management is one option. There's quantum factor investing. How how do you now ? How is what you know about quantum factory ? Westinghouse that affected how mercer looks attractive manages ? Sure. Well, actually, it's very thinking about factors is extremely important tool to apply to active managers. I mean, on using the tools and techniques that we'll have to look at those portfolios and understand where the true returns for amongst the fundamental active managers have come from her. So that provides is, um has done for some time a great opportunity to understand if the manager's generally adding alfa or is just pigging backing a factor. So so, so factor investing is helps raise the bar of those remaining high quality quality mean, standard, traditional, active manages traditional acting manages conversations with us. We can analyse their portfolios in terms of factors. We can explain the factors to them. If the results show that they've got some factor exposures or perhaps some signal exposures, we can confirm with them whether or not that makes sense in terms of their process on dh. If we show that they've got some unintended exposures in there, then that can also inform the process. I think the human brain is still full of pastie for surprises machines. Over the next few years, we will see more and more interesting ways off combining human insights with the systematic quantitative insights that we say from factor investing. Yes, manager skill, i believe, does does exist. So it's a it's a question of how you can potentially for a factor investor, how you combine them in our particular case, it didn't take us as i mentioned already. Our portfolio process is based on ri lansing exploiting this rebalancing of eight. We've developed performance attribution tools which can measure that signal and measure what the contribution to the return has bean from this portfolio rebounding activity, and then at the same time measure what the residual is, which we would view as being the influence of all the sort of extraneous elements in due to the portfolio positioning over time, those those dwindle away to nothing on what is left is thie what we view very much is alfa above and beyond what what may or may not have been contributed by factor explosions to quick question do you call the people on your investment ? Scheme fund managers or do you have a different title ? Waded portfolio manager phyllis ? We do indeed, this is very much an active on active process, even though its quantitative on dh systematic in nature, of course, the parameters that have been chosen, tio tio optimize the portfolios, how we rebounds, how often we rebounds the rules associated without the trading system. These have all been designed by us. This is sort of skill based, active management and requires skilled operators that we do call portfolio managers, and secondly, could you give us an example of something that's alfa that isn't systematic something where, you know, the individual out outside of this factory compost model couldn't pick that up ? Otherwise, whether that's luck or skill, we've got to put that in the alfa bucket. Well, i mean, this is this is this might be a sort of sack religious to the sort of supporters of factor investing she's going to be very hard because because the paper i was referring to earlier, it was interesting. So all of those all of those various different factor portfolios that had been created if if they weren't able to explain the performance by factors the call went out to find what is the missing factor that could be used to explain these these results so therefore it must everything must be attributable to affect a rather than the sort of dark matter that is, that is manager skill. So i would say, i would say that alfa does exist it's hard to define on maybe most has a has a view on this as well, but certainly the constant sort of rule rule the world is there a most definitional result for just the beater ? You can't explain what let me answer on important question, which is mercer believes in active management. We do believe that skill is that minute scarce it's hard to find. But we do believe that there is skill in our industry which can generate alfa above and beyond that explained by factor returns very quick for what does this all mean for fees ? I mean, if you can create a model with a lot of obviously a lot of skill and input that goes into it. But after that it's it's up s o i think i mentioned earlier in a systematic, transparent and low cost. These are very important two accidents on dh it is a downward pressure that's our experience is that the factory investing solutions generally offered a d a lower fee reiter and that is something we encourage on behalf of our clients. S so yes, it's disruptive to the to the hi fi model that some active manages employ. We're really excited about the opportunity on fees because what we think we can do with a scalable, systematic solution like venture investing solutions we provide is bring something better teo portfolio of all sorts of regular investors who couldn't otherwise afford to access that rare. We have to find how to access alfa that michael was talking about. So for defined contribution investors, they can't afford a lot of the time traditional active fees, and we can bring them a high quality product, which gives that little bit extra grand return over the course of their working life. It makes their time just get better by using factor investing, i would say that there is there is certainly downward pressure on fees this's clear across the whole industry. However there are investors still are prepared and willing to pay for consistency, and i think sort of passive implementations of factor base portfolios, which can undergo long periods of underperformance. They're very, very cyclical relative teo, rather, to the cap weighted index, which you know, even though it's, not an optimal portfolio, remains the sort of benchmark of choice for most people. Even if it's, not a formal benchmark, it's, at least provides the context by which these other equity portfolios judge. So we believe people are still prepared to pay for this sort of consistency of that performance, which perhaps the some of the passive implementations of factor explosions don't necessarily offer. And i want to move on in the final third of the program to house scaleable factor investing is if there's all this money going in does it mean it's all chasing the same stocks ? Michael, can you give us some idea of just how you think about this issue of crowding ? If everybody once valley was a factor, is there a danger ? Sure, whenever people without any type of investment strategy benefactor, otherwise you're thinking about the potential risks, you know, once got to be cognizant of that. And so the issue of crowding is something to be aware off. We look at take no volatilities that's often, you know example of factor, whether question crowding arises, so it must weigh follow a large number of low volatility factor products. Way get holdings from those managers on regular basis. We analyze those we can see how people are different, and so we dio do follow that quite closely in on a dynamic, ongoing basis. Say the risk for crowding is not the number one risk to think always gotta think about it, it's not the highest severity risk one that we see the moment based on the evidence of the holdings data that we get across cross available products, but it's certainly something people should not close their eyes to the risk of it. Yeah, i would agree with michael. Definitely. There is diversification to be had between different providers of these of these products and using low volatility. Azan example, just put some numbers on this. If you look at a sort of, if you look at this spread of returns within on active manager database in the low volatility field, this is now this is some research that we did between the period of oh five and on the end of twenty sixteen, the average difference between a top quarter performer in us low volatility and a third quarter performer was around six and a half percent, and for global equities, it was around about five percent. These have clearly very different outcomes that are being offered by managers offering the same thing. So there was a danger of crowding, if if if all people are doing is pursuing naive strategies, loading up on the least volatile stocks, which which anybody can measure. But there are more sophisticated and differentiated strategies out there that it's worth on. Important, in fact, i would say teo identify who was different, but to what extent was that difference in performance down to the fact they had different underlying stocks for stop or to what extent was it down to the fact they just had different exposures to the same stocks ? Because anyone who's invested and look a top ten holdings ? And you could get managed with very different performance records and they've all got shell and they've all got bp and they've all got unilever. A bit of both is the answer we don't have access to the other managers holding sure, yeah, but maybe the different way. Line them all up and have a look at them there's often commonality towards the top end of the portfolio, but thereafter, actually is a bit of an aside, i find interesting to think that a lot of love a love scene bility managed tend to have the perception that they might be uniquely different to everyone else because they're so different, you know, to the index, but but everyone's different index, and they're all crowded up in another part of the of the playing field, but nevertheless the dispersion between them. Is is evident from looking. I'd like to focus on a slightly different aspect from creating crowding kind of summons up sort of images of two thousand seven long short type against, i think that's a low risk outcome for the sorts of long only strategies which we've focused on here focus instead on capacity, meaning lots of money is coming into factor investing strategies and to some extent that will push up the prices today off the stocks that are associated that and therefore cushion and the future returns. So really to me this is a capacity issue more than a crowding issue. It's about long run expected returns muchmore than it's about some short, round, nasty event and for longer, unexpected returns on the impact of crowding there's really or capacity of the patent there's really two things to think about one is other signals which we're investing in us commoditized or not, i touched on that earlier, and michael on david have both been talking about that, but the other is what's the nature of the strategy itself and is the strategy itself designed effectively for almost limitless capacity, in which case is going to go into a very large stocks, too, or is it designed for some reasonable amount of capacity, in which case it'll go into smaller stocks ? And the reason i want to focus on this aspect of capacity is that there is overwhelming evidence that factor returns are richer in smaller stocks and largest ox. So if you're buying something that is designed for forty fifty plus billion dollars to go into it, the only way khun do that is by investing in a bunch of stocks which don't really reward factor exposures anyway, would you go ? I think, the point that i was going to add to what actually just is that the constraints that index index providers for factor exposures are forced, teo forced to use in order to make their in the indices investable in that large capacity immediately make them sort of suboptimal and gaining exposure to that to that factor. So, for example, needed an experiment based on the one of the went say which index provided, but one of the well known minimum variance indexes or minimum volatility index is on dh, which has a raft of constraints, which forces, which enables them to be investable in large quantity, but that doesn't necessarily make them as minimum variance is, they potentially could be on its possible to further reduce the minute. The volatility of the allegedly minimum variance index is quite easily if you don't have to be bullet with these constraints very quickly and then i'll bring michael. But have you got any strategies that you have ? Captain, just gentlemen, this is the most we can run in this otherwise performance. We will have to change what we do. We don't want to do that way. Have not yet had to do that. But we are very cognizant of capacity. Andi, we have a clear idea off what the capacity limitations are. The strategies it's extremely important. His in asset management history is littered with examples of managers that have taken on far too many assets than the strategy can handle. Asher, you will mention i think earlier about the growth things like pensions and dc. Is there a danger that the man in the street will end up with a sub optimal product ? Because, that's, what everyone's going to get put into big, big developed for i think the real risk in dc schemes is for schemes with limited governance budgets, a limited amount of time to spend digging into the investment proposition that something that can be described very easily as an index or very easily as a single factor is easy for trustees to swallow. But from an investment perspective really risks being suboptimal from all of the considerations we've talked about from signal inclusion through re balancing frequency through multi factor and portfolio construction. Michael, have you seen much evidence of providers of factor and quant models discussing capacity way have to have that conversation every time we meet with the money and manage money manager on its a very good idea to write down what they say on each different occasion as you go through time, it can creep up. I am no capacity is incredibly important azan aspect of crowding, andi also just that people run through those capacity limits. The product usually changes in some way and it's not what you originally built in the first place. So, you know, capacity management is very important consideration within implementation. So, yes, we talked a lot about you factors in kuantan alongside do you ever come across managers who say khun ? Reverse this. This is a brilliant way of shorting that's. Where what the the times has to think about their restore ince's. What those risks are on. Dh on shorting on time. You know the time horizon and ability to be patient and their belief in certain factors could be influenced in a different way if they have a short position to that in a long position so clearly or risk premium strategy is alternative risk premium strategies which do employ the long and the short party of the poor failure. Now, interestingly, sg three questions, eyes is sg affected what's your so yes, she is lots of different things, right ? At the very minimum, its environment, social and governance and none of those things they're uniquely defined. So it could be really careful about what we mean. I knew this was a bad questions whilst right at the end, my kind of broad take on it is actually it's really important. Yes, we see it being really important for client portfolios. We say a lot off. Very kind of ticked the box approaches to sg. We believe that the best approach is to incorporate a ll three of those. Areas as factors when one select stocks on that's absolutely what we do in our sustainable multi factor approach way measure the exposure of each stock in our investable universe, it's, environmental and social approaches and its governance approaches. And we capture in the portfolio those stocks in accordance with how will they do on those metrics ? Justus, we do value and quality. And how does that how does that impact is one thing ? Offices this shareholders you khun vote. So yes, how does that i mean, i suppose think of example, if you had a company that could do something that would really help the momentum factor if you like, we'll do something really help the quality factor wait even lightly. Different position. I think my my co panelists here that there's, an investment management company, we do no fundamental analysis of companies. What so ever wear sort of agnostic on these on these on these issues which doesn't mean to say that we have not been able to incorporate sg elements in trouble follows at the request of our clients, we have found actually that incorporating yes, he scores of saying support for the optimization to boost the school residue, that of the index is something that is to use michael's earlier technical word orthogonal to what were the three of us have set out what we're doing. So we are mutual on whether or not sg fractures will improve or reduce return. However, we have found that it has no impact on on the ability to generate to generate alfa or control risk. So we have been able to incorporate these elements without taking a view on them, which is to remain true to our sort of investments and vote on revolution. If we have been delegated authority to vote, we actually outsource. We always vote. The voting is that sourced. We offer our investors the choice to select from the menu of voting options, one of which is sg friendly and many choose that. What would you say ? That sounds more like an absolute landlord that an investor they obviously seeing this stock just as a piece of paper that throws ofsome different economic signals. It's a business it's got people in there. She said that that's true, we're trying teo generate returns for r r, for our clients in a risk controlled way and the way that we have to do this is has been developed over thirty years using mathematics rather than fundamental analysis. It's not all skill set were were obviously aware that these are companies that do business in the real world, but we're generating returns from from other sources. So it's of less importance, we've much time want to get a final thought from all of you recovered a lot in the last forty minutes or so. David, i come to you first to this one key message around factor or factors type investing. If it's has come out of this, what would it be if i would say ? Don't forget the importance of diversification diversifying between factors on dh looking beyond factors for other un correlated alfa sources that aren't factors, diversification is sometimes thought of eyes the only free lunch and investing its not a tall of free lunch diversified allocation needs to be maintained on bat cost money, but don't forget to diversify to this. Great actually, i would say i think the fundamentals of point to remember is that factor investing and indeed the reinvesting i really fast developing progressing fields wear the quality of what you put into the portfolio in mexico difference to the results we get from the portfolio. So look for solutions that are high quality and are looking to improve as we learn more and more about the world. Thank you, transparency and low cost. But we have to go there, gentlemen. Thank you all very much, indeed, for joining us. Thank you for watching from all of us here. Goodbye for now.